Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

gyroscopes and LP filter

Status
Not open for further replies.
S

senmeis

Full Member level 3
Joined
Nov 26, 2014
Messages
160
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,296
Activity points
2,473
Hi,

as far as I know some gyroscopes have an interna LP filter at the output such as https://www.systron.com/sites/default/files/964001_l-qrs11_3.pdf: Bandwidth >60 Hz.

Questions:
1. Why LP filter? What if no LP filter?
2. What’s the meaning of „-90 phase shift“?
3. Are LP filters generally used for other types of sensors?

Senmeis
 

The LP filter is used to filter out high frequency noise. It is commonly used in many sensor applications. But I have no idea what that -90 degree thing means.
 

There are two motivations for the low pass:
- reduce the gyroscope signal noise bandwidth and respective variance
- filter out residuals of gyroscope carrier frequency

Low pass filters have a magnitude and phase characteristic. Specifying phase without magnitude is unusual and - in my view - somehow useless.
 

Hi,

as far as I know some gyroscopes have an interna LP filter at the output such as https://www.systron.com/sites/default..._l-qrs11_3.pdf: Bandwidth >60 Hz.

Questions:
1. Why LP filter? What if no LP filter?
2. What’s the meaning of „-90 phase shift“?
3. Are LP filters generally used for other types of sensors?
Click to expand...

I wonder .. because there is nothing mentioned given in the datasheet.
Is this just assumption or did you read it somewhere? If you read it show us where.

Klaus
 

KlausST said:
Hi,
I wonder .. because there is nothing mentioned given in the datasheet.
Is this just assumption or did you read it somewhere? If you read it show us where.
Klaus
Click to expand...

Yes - I could not find any mentioning of -90 deg....
 

as far as I know some gyroscopes have an interna LP filter at the output such as https://www.systron.com/sites/default..._l-qrs11_3.pdf: Bandwidth >60 Hz.
Click to expand...
Some responses are more than intuitive, as for example filtering signals above 60Hz suggests eliminating the effect of vibration from loose (not glued) slides on laminated transformers, although gyroscopes are intended to be embeeded (where not often the power supply is AC) . By the way, gyroscopes, unlike accelerometers, are not designed to give quick response, but to act as a baseline for steering direction, and this may explain to you why having low pass filtering.
 

as for example filtering signals above 60Hz suggests eliminating the effect of vibration from loose slides on transformers...
Click to expand...

But the specs say >60Hz (I understand that it means low frequencies are removed and high frequencies are passed on to the output...)

What I am missing?
 

I understand that it means low frequencies are removed and high frequencies are passed on to the output...
Click to expand...
It certainly doesn't mean this, just say the low pass bandwidth is greater than 60 Hz.

I could not find any mentioning of -90 deg....
Click to expand...

90degree.PNG
 

HI,

(I understand that it means low frequencies are removed and high frequencies are passed on to the output...)
Click to expand...

I think it means bandwidth: from 0 .. 60Hz
The bandwidth specification has tolerance. ">60Hz" just says the the upper bandwith limit is rather 63Hz than 59Hz.

Klaus
 

senmeis said:
So it’s a bit strange to use „-90deg phase shift“?
Click to expand...

This is just a reference of the delay generated by the output filter; depending on how instantly you want the sensor response, that information become relevant, otherwise for relatively steady systems in space (with a bulky mass), where consecutive readings would give small shift in current orientation, the phase shift doesn`t matter at all.
 

So it’s a bit strange to use „-90deg phase shift“?...
Click to expand...

I do not understand why you find the phase shift strange: filters DO produce phase shifts and 90 deg phase shift is the 1/2 output point...
 
Last edited by a moderator:

filters DO produce phase shifts and 90 deg phase shift is the 1/2 output point...
Click to expand...
Why "1/2 output", that's just a speculation without knowing the magnitude and phase characteristic. The relation will be different for each filter order and steepness. In case of first order, -90 degree is e.g. the asymptotic phase for f >> fc.

It's not strange that the filter has -90 degree at some frequency, it's just a strange specification. At least if nobody can derive that -90 degree phase shift is a characteristic parameter of the gyroscope characteristic.
 

Since the unit only has one analog output and phase is a relative measurement, it has no meaning as far as the gyroscope reading is concerned so I assume it is just their way of describing the LPF characteristic. With no internal block diagram we cannot tell what kind of filter it uses but to describe in terms of phase implies it is a simple active or passive RC type rather than a digital one. It could probably be described in terms of dB roll off equally well.

Whatever it means, I'm sure it is there to filter out noise, internal reference signals and maybe to calm jitter in the output waveform.

Brian.
 

Hi,

Since the unit only has one analog output and phase is a relative measurement, it has no meaning
Click to expand...
I don´t agree.

Often those gyroscopes are use for mechanical stabilisation.
For small movements piezos are used, then magnetics, up to stepper motors or BLDC motors.
The gyroscopes build a feedbacked loop with the control unit, the motors and the mechanics.

The loop stability is determined by phase shift (and gain). Thus the phase shift plays a major role to determine a system´s stability and upper usable frequency limit.
For a detailed analysis one needs the phase and gain chart.

Klaus
 

Why "1/2 output", that's just a speculation without knowing the magnitude and phase characteristic. The relation will be different for each filter order and steepness...
Click to expand...

Of course you are right and I was careless. 1/2 output commonly corresponds to the 3db attenuation point widely used for slope reference. For a first order filter, the phase shift is -90 at the cutoff frequency.

I guess the manufacturer just wanted to specify the cut-off value for the filter because user should stay away from this point...

- - - Updated - - -

Since the unit only has one analog output and phase is a relative measurement...
Click to expand...

I am not so sure but I guess the phase is relative to the motion (pitch and roll) and in the datasheet (https://www.analog.com/media/en/technical-documentation/data-sheets/ADXRS290.pdf) they mention an input reference...
 

A second order Butterworth filter has -90 degree phase at the cut-off frequency (-3 dB point).
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top